Applications-driven parallel I/O

N. Galbreath, W. Gropp, D. Levine

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We investigate the needs of some massively parallel applications running on distributed-memory parallel computers at Argonne National Laboratory and identify some common paralell I/O operations. For these operations, routines were developed that hide the details of the actual implementation (such as the number of parallel disks) from the application, while providing good performance. An important feature is the ability for the application programmer to sspecify that a file be accessed either as a high-performance parallel file or as a conventional Unix file, simply by changing the value of a parameter on the file open call.

Original languageEnglish (US)
Title of host publicationProceedings of the Supercomputing Conference
PublisherPubl by IEEE
Pages462-471
Number of pages10
ISBN (Print)0818643404, 9780818643408
DOIs
StatePublished - 1993
Externally publishedYes
EventProceedings of the Supercomputing '93 Conference - Portland, OR, USA
Duration: Nov 15 1993Nov 19 1993

Publication series

NameProceedings of the Supercomputing Conference
ISSN (Print)1063-9535

Other

OtherProceedings of the Supercomputing '93 Conference
CityPortland, OR, USA
Period11/15/9311/19/93

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Applications-driven parallel I/O'. Together they form a unique fingerprint.

Cite this